INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

BuyableInstoreAndOnline

-0.85

çĶ°

-0.80

Db

-0.77

Eva

-0.73

Jenn

-0.73

Ke

-0.71

Da

-0.70

handle

-0.69

lip

-0.68

chy

-0.66

POSITIVE LOGITS

owship

0.69

oros

0.67

 propri

0.64

 contingency

0.63

 tension

0.62

ativity

0.60

 stewards

0.59

onet

0.59

 believers

0.58

 province

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.